Atlas

Benchmarks

← All benchmarks

Arena Text — Software & IT Services (No Style Control)

Professional Work

Raw Arena rating for software and IT-services Text Arena prompts without style-control adjustment. Higher ratings indicate stronger preference in blind pairwise user votes.

Top models (higher is better)

ModelScore
Gemini 4 Argon1544
Claude Opus 5.51538
Claude Fable 5.11529
Claude Opus 51528
Opus 4.61528
MiMo-V2.6-Pro1527
Opus 4.71516
Muse Wonder1514
Claude Fable 51512
Muse Spark 1.31511
Gemini 3.8 Flash1506
Muse Spark 1.21506
Claude Sonnet 5.51501
GLM-5.3 Flash1501
Gemini 3.7 Flash1499
Qwen3.8-Max1498
Kimi K31498
Muse Spark 1.11498
MiMo-V2.6-Flash1497
DeepSeek V4.1 Flash1496
Qwen3.7 Max Preview1495
MiMo-V2.5-Pro1494
GLM-5.31494
Gemini 3.5 Flash1494
GPT-5.51493
Gemini 3.6 Flash1492
ERNIE 5.11492
GPT-5.41491
Sonnet 4.61491
GPT-5.6 Sol1489
Opus 4.81488
Gemini 3.1 Pro Preview1487
Muse Spark1487
GLM-5.21486
Gemini 3 Pro Preview1485
GLM-5.11484
Kimi K2.61484
Opus 4.51483
GPT-5.6 Terra1482
GPT-6.1 Sol1479
Loading Atlas data…